Data is never just data

The need to look harder

The site displays statistics over a 3-month period because the data is less important in itself than in how it adds up to a trend. That trend conventionally is shown as mapped against the now-famous and desired flattened COVID curve. How closely your town’s curve overlays it is a sign of the town’s success and ofyour risk. The curve serves an important governmental role as well: If the curve has not been flattened for some number of weeks, your town should not be reopening its local economy.

The town displays data about the breakdown of COVID-19 cases by race. In this way, it has made a decision not only that race is significant but also about what constitutes a race. We can see that Black people account for 8% of the COVID-19 cases in this town, whereas 67% of the COVID patients are white. But we have to look much harder and go to another page to find out that only 3.17–4.19% of the residents are Black; it’s a range because 4.19% of residents are of two races. Either way, the percentage is shameful.

Without knowing the racial breakdown of the population overall, the 8% figure is accurate but misleading. The incidents of COVID-19 by race are included presumably because we want to know if this is an equal opportunity virus. Without knowing the population’s breakdown by race, we cannot see that the disease has hit Black residents about two to three times as hard as white residents.

Data is never simple

So, is that 8% figure data in this context? Yes. Is it misleading in this context? Yes. That makes it evidence that data is never just data; it is only data within conventions of measurement. It is only data because we chose to generate it. That purpose is embedded in a web of assumptions, intentions, history, and values. Data is never simple, is never isolated, and is never the whole story in and of itself.

We need to heed the data—in this era of pandemic more than ever. But we should not think that data is the simple truth.